ben/mgmt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

111

Browse Source
This commit is contained in:
Houzhong Xu
2025-10-09 01:22:22 +00:00
parent 1c994f9f60
commit eab95c8c80
136 changed files with 11001 additions and 849 deletions
Download Patch File Download Diff File Expand all files Collapse all files

110
components/nomad/jobs/install-podman-driver.nomad
View File

@@ -1,110 +0,0 @@
job "install-podman-driver" {
datacenters = ["dc1"]
type = "system" # 在所有节点上运行
group "install" {
task "install-podman" {
driver = "exec"
config {
command = "bash"
args = [
"-c",
<<-EOF
set -euo pipefail
export PATH="/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin"
# 依赖工具
if ! command -v jq >/dev/null 2>&1 || ! command -v unzip >/dev/null 2>&1 || ! command -v wget >/dev/null 2>&1; then
echo "Installing dependencies (jq unzip wget)..."
sudo -n apt update -y || true
sudo -n apt install -y jq unzip wget || true
fi
# 安装 Podman（若未安装）
if ! command -v podman >/dev/null 2>&1; then
echo "Installing Podman..."
sudo -n apt update -y || true
sudo -n apt install -y podman || true
sudo -n systemctl enable podman || true
else
echo "Podman already installed"
fi
# 启用并启动 podman.socket，确保 Nomad 可访问
sudo -n systemctl enable --now podman.socket || true
if getent group podman >/dev/null 2>&1; then
sudo -n usermod -aG podman nomad || true
fi
# 安装 Nomad Podman 驱动插件（始终确保存在）
PODMAN_DRIVER_VERSION="0.6.1"
PLUGIN_DIR="/opt/nomad/data/plugins"
sudo -n mkdir -p "${PLUGIN_DIR}" || true
cd /tmp
if [ ! -x "${PLUGIN_DIR}/nomad-driver-podman" ]; then
echo "Installing nomad-driver-podman ${PODMAN_DRIVER_VERSION}..."
wget -q "https://releases.hashicorp.com/nomad-driver-podman/${PODMAN_DRIVER_VERSION}/nomad-driver-podman_${PODMAN_DRIVER_VERSION}_linux_amd64.zip"
unzip -o "nomad-driver-podman_${PODMAN_DRIVER_VERSION}_linux_amd64.zip"
sudo -n mv -f nomad-driver-podman "${PLUGIN_DIR}/"
sudo -n chmod +x "${PLUGIN_DIR}/nomad-driver-podman"
sudo -n chown -R nomad:nomad "${PLUGIN_DIR}"
rm -f "nomad-driver-podman_${PODMAN_DRIVER_VERSION}_linux_amd64.zip"
else
echo "nomad-driver-podman already present in ${PLUGIN_DIR}"
fi
# 更新 /etc/nomad.d/nomad.hcl 的 plugin_dir 设置
if [ -f /etc/nomad.d/nomad.hcl ]; then
if grep -q "^plugin_dir\s*=\s*\"" /etc/nomad.d/nomad.hcl; then
sudo -n sed -i 's#^plugin_dir\s*=\s*\".*\"#plugin_dir = "/opt/nomad/data/plugins"#' /etc/nomad.d/nomad.hcl || true
else
echo 'plugin_dir = "/opt/nomad/data/plugins"' | sudo -n tee -a /etc/nomad.d/nomad.hcl >/dev/null || true
fi
fi
# 重启 Nomad 服务以加载插件
sudo -n systemctl restart nomad || true
echo "Waiting for Nomad to restart..."
sleep 15
# 检查 Podman 驱动是否被 Nomad 检测到
if /usr/local/bin/nomad node status -self -json 2>/dev/null | jq -r '.Drivers.podman.Detected' | grep -q "true"; then
echo "Podman driver successfully loaded"
exit 0
fi
echo "Podman driver not detected yet, retrying once after socket restart..."
sudo -n systemctl restart podman.socket || true
sleep 5
if /usr/local/bin/nomad node status -self -json 2>/dev/null | jq -r '.Drivers.podman.Detected' | grep -q "true"; then
echo "Podman driver successfully loaded after socket restart"
exit 0
else
echo "Podman driver still not detected; manual investigation may be required"
exit 1
fi
EOF
]
}
resources {
cpu = 200
memory = 256
}
// 以root权限运行
// user = "root"
# 使用 nomad 用户运行任务，避免客户端策略禁止 root
user = "nomad"
# 确保任务成功完成
restart {
attempts = 1
interval = "24h"
delay = "60s"
mode = "fail"
}
}
}
}

43
components/nomad/jobs/juicefs-controller.nomad Normal file
View File

@@ -0,0 +1,43 @@
job "juicefs-controller" {
datacenters = ["dc1"]
type = "system"
group "controller" {
task "plugin" {
driver = "podman"
config {
image = "juicedata/juicefs-csi-driver:v0.14.1"
args = [
"--endpoint=unix://csi/csi.sock",
"--logtostderr",
"--nodeid=${node.unique.id}",
"--v=5",
"--by-process=true"
]
privileged = true
}
csi_plugin {
id = "juicefs-nfs"
type = "controller"
mount_dir = "/csi"
}
resources {
cpu = 100
memory = 512
}
env {
POD_NAME = "csi-controller"
}
}
}
}

38
components/nomad/jobs/juicefs-csi-controller.nomad Normal file
View File

@@ -0,0 +1,38 @@
job "juicefs-csi-controller" {
datacenters = ["dc1"]
type = "system"
group "controller" {
task "juicefs-csi-driver" {
driver = "podman"
config {
image = "juicedata/juicefs-csi-driver:v0.14.1"
args = [
"--endpoint=unix://csi/csi.sock",
"--logtostderr",
"--nodeid=${node.unique.id}",
"--v=5"
]
privileged = true
}
env {
POD_NAME = "juicefs-csi-controller"
POD_NAMESPACE = "default"
NODE_NAME = "${node.unique.id}"
}
csi_plugin {
id = "juicefs0"
type = "controller"
mount_dir = "/csi"
}
resources {
cpu = 100
memory = 512
}
}
}
}

55
components/nomad/jobs/nomad-consul-config.nomad
View File

@@ -1,55 +0,0 @@
job "nomad-consul-config" {
datacenters = ["dc1"]
type = "system"
group "nomad-server-config" {
constraint {
attribute = "${node.unique.name}"
operator = "regexp"
value = "semaphore|ash1d|ash2e|ch2|ch3|onecloud1|de"
}
task "update-nomad-config" {
driver = "exec"
config {
command = "sh"
args = [
"-c",
"sed -i '/^consul {/,/^}/c\\consul {\\n address = \"master.tailnet-68f9.ts.net:8500,ash3c.tailnet-68f9.ts.net:8500,warden.tailnet-68f9.ts.net:8500\"\\n server_service_name = \"nomad\"\\n client_service_name = \"nomad-client\"\\n auto_advertise = true\\n server_auto_join = true\\n client_auto_join = false\\n}' /etc/nomad.d/nomad.hcl && systemctl restart nomad"
]
}
resources {
cpu = 100
memory = 128
}
}
}
group "nomad-client-config" {
constraint {
attribute = "${node.unique.name}"
operator = "regexp"
value = "master|ash3c|browser|influxdb1|hcp1|warden"
}
task "update-nomad-config" {
driver = "exec"
config {
command = "sh"
args = [
"-c",
"sed -i '/^consul {/,/^}/c\\consul {\\n address = \"master.tailnet-68f9.ts.net:8500,ash3c.tailnet-68f9.ts.net:8500,warden.tailnet-68f9.ts.net:8500\"\\n server_service_name = \"nomad\"\\n client_service_name = \"nomad-client\"\\n auto_advertise = true\\n server_auto_join = false\\n client_auto_join = true\\n}' /etc/nomad.d/nomad.hcl && systemctl restart nomad"
]
}
resources {
cpu = 100
memory = 128
}
}
}
}

23
components/nomad/jobs/nomad-consul-setup.nomad
View File

@@ -1,23 +0,0 @@
job "nomad-consul-setup" {
datacenters = ["dc1"]
type = "system"
group "nomad-config" {
task "setup-consul" {
driver = "exec"
config {
command = "sh"
args = [
"-c",
"if grep -q 'server.*enabled.*true' /etc/nomad.d/nomad.hcl; then sed -i '/^consul {/,/^}/c\\consul {\\n address = \"master.tailnet-68f9.ts.net:8500,ash3c.tailnet-68f9.ts.net:8500,warden.tailnet-68f9.ts.net:8500\"\\n server_service_name = \"nomad\"\\n client_service_name = \"nomad-client\"\\n auto_advertise = true\\n server_auto_join = true\\n client_auto_join = false\\n}' /etc/nomad.d/nomad.hcl; else sed -i '/^consul {/,/^}/c\\consul {\\n address = \"master.tailnet-68f9.ts.net:8500,ash3c.tailnet-68f9.ts.net:8500,warden.tailnet-68f9.ts.net:8500\"\\n server_service_name = \"nomad\"\\n client_service_name = \"nomad-client\"\\n auto_advertise = true\\n server_auto_join = false\\n client_auto_join = true\\n}' /etc/nomad.d/nomad.hcl; fi && systemctl restart nomad"
]
}
resources {
cpu = 100
memory = 128
}
}
}
}

34
components/nomad/jobs/nomad-nfs-volume.nomad
View File

@@ -1,34 +0,0 @@
job "nfs-volume-example" {
datacenters = ["dc1"]
type = "service"
group "nfs-app" {
count = 1
volume "nfs-shared" {
type = "host"
source = "nfs-shared"
read_only = false
}
task "app" {
driver = "podman"
config {
image = "alpine:latest"
args = ["tail", "-f", "/dev/null"]
}
volume_mount {
volume = "nfs-shared"
destination = "/shared"
read_only = false
}
resources {
cpu = 100
memory = 64
}
}
}
}

43
components/nomad/volumes/nfs-csi-volume.hcl Normal file
View File

@@ -0,0 +1,43 @@
# NFS CSI Volume Definition for Nomad
# 这个文件定义了CSI volume，让NFS存储能在Nomad UI中显示
volume "nfs-shared-csi" {
type = "csi"
# CSI plugin名称
source = "csi-nfs"
# 容量设置
capacity_min = "1GiB"
capacity_max = "10TiB"
# 访问模式 - 支持多节点读写
access_mode = "multi-node-multi-writer"
# 挂载选项
mount_options {
fs_type = "nfs4"
mount_flags = "rw,relatime,vers=4.2"
}
# 拓扑约束 - 确保在有NFS挂载的节点上运行
topology_request {
required {
topology {
"node" = "{{ range $node := nomadNodes }}{{ if eq $node.Status "ready" }}{{ $node.Name }}{{ end }}{{ end }}"
}
}
}
# 卷参数
parameters {
server = "snail"
share = "/fs/1000/nfs/Fnsync"
}
}

22
components/nomad/volumes/nfs-dynamic-volume.hcl Normal file
View File

@@ -0,0 +1,22 @@
# Dynamic Host Volume Definition for NFS
# 这个文件定义了动态host volume，让NFS存储能在Nomad UI中显示
volume "nfs-shared-dynamic" {
type = "host"
# 使用动态host volume
source = "fnsync"
# 只读设置
read_only = false
# 容量信息（用于显示）
capacity_min = "1GiB"
capacity_max = "10TiB"
}

22
components/nomad/volumes/nfs-host-volume.hcl Normal file
View File

@@ -0,0 +1,22 @@
# NFS Host Volume Definition for Nomad UI
# 这个文件定义了host volume，让NFS存储能在Nomad UI中显示
volume "nfs-shared-host" {
type = "host"
# 使用host volume
source = "fnsync"
# 只读设置
read_only = false
# 容量信息（用于显示）
capacity_min = "1GiB"
capacity_max = "10TiB"
}